The Service "kube-dns" is invalid: spec.clusterIP: Invalid value: "10.96.0.10": provided IP is not in the valid range. The range of valid IPs is 10.0.0.0/24 #2240
Comments
|
👍 |
|
cc @r2d4 |
I did some tests involving different iso files (see attachements). Approach 1 & 2 failed. So I replaced the new minikube.exe with the old one (approach 3). My tests started to work again. :) I compared the log of the installation and noticed that different flags are passed:
InsecureRegistry:[10.0.0.0/24] differs. Of course, this could be coincidence, but perhaps it's helpful. Let me know if you need further logs. |
|
This will be fixed by #2245. We will be getting a patched release out today. In the meantime, there are a few different workarounds:
Of course, once we release a patch version, you won't need to use any of these. |
|
@r2d4 I had to use one of the workarounds (I chose the first one) to get v0.24.1 working on Mac. I'm assuming that's not expected? |
|
Now there's something weird with my Mongo service using the kubeadm bootstrapper, not setting itself to master. Could be a communication issue with the sidecar container in the same pod.....don't know yet. When I revert back to v0.23.0, everything is fine. I'll try to come up with a recreatable test.....but this doesn't seem to be working for Mac. Here's what I've tried so far:
|
|
After some testing, here's a little more information about the above. All those example were done using kubernetes 1.7.5. When I move to kubernetes 1.8.0, everything I mentioned above works. So it's some incompatibility with minikube 0.24.1 and kubernetes 1.7.5. We're a little stuck on 1.7.5 until this issue gets fixed: |
|
Sorry, seems this was brought up in this open issue already: |
Minikube version (use
minikube version): minikube version: v0.24.0cat ~/.minikube/machines/minikube/config.json | grep DriverName): hypervcat ~/.minikube/machines/minikube/config.json | grep -i ISOorminikube ssh cat /etc/VERSION): minikube-v0.23.5.isoWhat happened:
Updated to 0.24 from 0.23 and kube-dns is logging errors. Hostname resolving fails (see below for further information)
What you expected to happen: No errors are logged and the hostname resolving within pods works as in 0.23.
How to reproduce it (as minimally and precisely as possible):
Simply install minikube by calling
minikube start --vm-driver="hyperv" --hyperv-virtual-switch="vSwitchExt" --v=3 --alsologtostderr. Deploy the two pods like in 0.23. Call first one.Output of minikube logs (if applicable):
by request
Anything else do we need to know:
Hi there,
I had to reinstall my minikube-instance, so I scrubbed any ".minikube" and ".kube" directories in addtion to the vm.
My test case did run in the previous version and now fails due to UnknownHostExceptions. Digging in the logs I found an error output (see below) and googled for kube-dns is invalid... which lead me to the issue #1908 .
Did something go wrong while installing the "new" instance or might this be broken in the new release? Since I'm a beginner with Kubernetes/Minikube, let me know if you need additional information (and how to get them ;))
The text was updated successfully, but these errors were encountered: